Model comparison
Llama 3.1-70B vs Sonar
Sonar is the stronger model overall, scoring 38.5 to 29.6 on the Noometry Index. Llama 3.1-70B costs 2.5× less per token, which makes it the better buy when Sonar's lead doesn't matter for your workload.
Last verified . 0 shared benchmarks.
Meta
PerplexitySummary
- The widest gap is in math, where Sonar leads 33.7 to 13.5.
- Llama 3.1-70B is cheaper at $0.40 / $0.40 per million input/output tokens, against $1 / $1 for Sonar.
- Llama 3.1-70B has downloadable open weights; the other is API-only.
Side by side
| Llama 3.1-70B | Sonar | |
|---|---|---|
| Provider | Meta | Perplexity |
| Noometry Index | 29.6 | 38.5 |
| Released | 2024-07-23 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | 128K | 128K |
| Max output | 4K | 4K |
| Input $ / M tokens | $0.40 | $1 |
| Output $ / M tokens | $0.40 | $1 |
| Results tracked | 35 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Sonar leads
Llama 3.1-70B: 30.3 (#296), Sonar: 35.7 (#221)
| Benchmark | Llama 3.1-70B | Sonar |
|---|---|---|
| WeirdML | 9% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1260 | — |
| BigCodeBench Complete | 54.8% | — |
Agentic & Tool Use Not comparable
Llama 3.1-70B: 25.1 (#112), Sonar: —
| Benchmark | Llama 3.1-70B | Sonar |
|---|---|---|
| TheAgentCompany | 6.9% | — |
| BALROG | 27.9% | — |
Reasoning Too close to call
Llama 3.1-70B: 21.6 (#220), Sonar: 21.1 (#227)
| Benchmark | Llama 3.1-70B | Sonar |
|---|---|---|
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1241 | — |
| DTBench | 60% | — |
| LiveBench Data Analysis | — | 37.9% |
| LMCA | 14.8% | — |
| Epoch Capabilities Index | 125.92 | — |
| LiveBench | — | 46.9% |
Math Sonar leads
Llama 3.1-70B: 13.5 (#304), Sonar: 33.7 (#200)
| Benchmark | Llama 3.1-70B | Sonar |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 3.6% | — |
| Omni-MATH | 21% | — |
| LiveBench Math | — | 41.6% |
| LMArena Math | 1252 | — |
| MATH Level 5 | 36.7% | — |
Knowledge Not comparable
Llama 3.1-70B: 24.2 (#269), Sonar: —
| Benchmark | Llama 3.1-70B | Sonar |
|---|---|---|
| GPQA Diamond | 44.2% | — |
| MMLU-Pro | 65.3% | — |
| GPQA (HELM) | 42.6% | — |
| LMArena Expert | 1209 | — |
| MMLU | 80.1% | — |
Multilingual Not comparable
Llama 3.1-70B: 38.8 (#225), Sonar: —
| Benchmark | Llama 3.1-70B | Sonar |
|---|---|---|
| LMArena Non-English | 1219 | — |
| LMArena Chinese | 1215 | — |
| LMArena French | 1261 | — |
| LMArena German | 1222 | — |
| LMArena Japanese | 1132 | — |
| LMArena Korean | 1140 | — |
| LMArena Russian | 1234 | — |
| LMArena Spanish | 1253 | — |
Instruction Following Sonar leads
Llama 3.1-70B: 65.3 (#223), Sonar: 71.4 (#150)
| Benchmark | Llama 3.1-70B | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| IFEval | 82.1% | — |
| LMArena Instruction Following | 1231 | — |
Long Context Not comparable
Llama 3.1-70B: 37.6 (#214), Sonar: —
| Benchmark | Llama 3.1-70B | Sonar |
|---|---|---|
| LMArena Longer Query | 1241 | — |
Writing & Preference Sonar leads
Llama 3.1-70B: 35.4 (#267), Sonar: 52.6 (#167)
| Benchmark | Llama 3.1-70B | Sonar |
|---|---|---|
| LMArena Text | 1261 | — |
| LMArena Creative Writing | 1232 | — |
| EQ-Bench Creative Writing | 784 | — |
| WildBench | 75.8% | — |
| LMArena Multi-Turn | 1256 | — |
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is Llama 3.1-70B better than Sonar?
Sonar is the stronger model overall, scoring 38.5 to 29.6 on the Noometry Index. Llama 3.1-70B costs 2.5× less per token, which makes it the better buy when Sonar's lead doesn't matter for your workload.
Which is cheaper, Llama 3.1-70B or Sonar?
Llama 3.1-70B is cheaper. It lists at $0.40 per million input tokens and $0.40 per million output tokens; Sonar lists at $1 and $1.
Is Llama 3.1-70B or Sonar better for coding?
Sonar scores higher on coding benchmarks: 35.7 versus 30.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 128K tokens.
How many benchmarks do Llama 3.1-70B and Sonar share?
0 benchmarks have published results for both models. Llama 3.1-70B has 35 scored results on Noometry and Sonar has 7.